3.4 CLASSIFICATION OF UNWANTED SITUATIONS
3.4.1 High-price
3.4.1.1 The relation between high price and vulnerability
There are some methodical difficulties in the treatment of high prices and their relation with vul-nerability. If electricity is regarded as just another good, there should be no reason why high prices for this good would give special reasons for concern. However, electricity has some special characteristics that distinguish it from other goods:
• It is generally regarded as a necessity
• In Norway and to some extent Sweden it represents a significant share of some households’
expenditure
• At least in latter years, price variations have become relatively large
In this context, we shall not forget the underlying rationale for a market-based organization of the power sector, which is to increase economic efficiency. In a market, supply and demand adjust dynamically to the market price. Fluctuating prices are therefore not something “bad” that has to be avoided, but a necessary element in a well-functioning market. Of course, extreme prices can be problematic and may indicate deficiencies in the market structure. But occasional moderately high prices are a natural ingredient in markets, especially markets with large variations in both supply and demand, like the Nordic electricity market.
In [10] an analysis is made of the macro economic consequences of the high prices for electricity in 2003. The analysis uses the macro economic model KVARTS, which has a quarterly resolu-tion. Consequences are identified on:
• Norway’s real income
• Households’ disposable income, consume and saving rate
• Investments
• Export and import
• Gross domestic product (GDP)
• Employment
• Consumer price index (CPI)
The CPI increased with 1.3 percentage point for the year as a whole. Measured in fixed prices, households’ consumption was reduced with 0.3 percent, while households’ total consumption expenditure increased with 0.9 percent. The saving rate was reduced with 0.6 percentage point.
In addition to the effect of reduced demand from households, the competitiveness of export-oriented industries is reduced because of increased costs. Together with the direct effect of low
12X333 TR F5962
power production and reduced household demand, this reduces GDP of Norway (excluding the oil sector) with 0.5 percent, measured in fixed year-2000 prices. The direct contribution of reduced power production constitutes one third of this. Disposable real income of Norway is reduced with 6.6 billion NOK (0.8 billion Euros) or 0.5 percent. There is only a marginal effect on employment [10].
To our knowledge, no similar analysis has been done for the other Nordic countries. Even if the effects on households’ disposable income were similar, the other effects on the economy may be quite different. Denmark, for instance, would have a considerable increase in power production for export to Norway and Sweden. So while Danish consumers faced more or less the same prices as in Norway, Danish producers profited hugely from export revenues, which probably resulted in an increase of Danish GDP.
Macro economic analyses typically consider total amounts, and disregard distributional effects. In another study [11] the authors analyze how various groups of households were affected by the price increases. They divide households in ten income groups of equal size. As expected, there is a positive correlation between income and electricity consumption. As a result, the average in-crease in electricity expenses is calculated to 3172 NOK (386 Euros) per household per year for the lowest income group, and 5859 NOK (714 Euros) for the highest income group. However, there are a number of households in the lowest income groups that have a relatively high use of electricity. E.g. 17 percent of the lowest income group has an annual consumption in excess of 25000 kWh, and for a number of these households, high prices like in 2003 are problematic. On the other hand, many in these groups have various forms of support, which compensates the price increases to some extent.
In a study of the development in the power market in 2002-03 [12], the authors conclude that, looking back, the Nordic market handled the challenges of inflow shortage and resulting high price periods satisfactory. Based on the discussion above, the main problem of high prices, seems to be the position of low-income groups, especially those with high electricity consumption. Al-though many of these are support by various support schemes, some will not satisfy the criteria for these schemes.
So far, we have discussed the effect of high prices resulting from energy shortage. High prices can also occur as a result of a shortage of generation and/or import capacity during extreme cold and resulting high demand. In this case, the duration of high prices is probably only a few hours, pos-sibly several days in a row. E.g. on the day with the highest prices in Nordpool so far, 5 February 2001, had 8 hours with prices above 100 €/MWh. The level of such prices may become much higher. Nordpool has a technical price cap of approximately 1200 €/MWh, but this can probably be increased in special situations. Higher prices can also occur in the Balancing Market. In Nor-way, the price cap in this market is twice the price in Elspot, but at least 50000 NOK/MWH (appr.
6000 €/MWh). The question is to what extent this is a problem and if the possibility of very high prices during short periods represents a source of vulnerability in the Nordic system.
12X333 TR F5962
Presently only a small number of final consumers are directly exposed to the hourly spot price.
Most of these are large consumers with high competence on their energy consumption. So this form of high prices does not present a problem for consumers. However, a study by Svenska Kraftnät in 2002 points out that this constitutes a considerable risk for traders and especially re-tailers [13]. Thus there is a certain vulnerability of the market itself, if several market participants would go bankrupt in such cases.
3.4.1.2 Country-specific effects
The economic effects of high prices depend on a number of country-specific characteristics. Elec-tricity consumption as a share of total energy consumption gives an indication of the importance of electricity in a country’s energy consumption, and hence of the relative effect of high prices on the economy as a whole. In the previous Section, we referred to several analyses from the Norwe-gian SSB. The main impression is that the high prices in 2003 were not a major problem for the Norwegian economy as a whole, but that certain low-income households with high electricity consumption probably were severely affected. This fact and the political turmoil it causes may be the worst effect of high prices, and therefore it is important to focus on characteristics that high-light the importance of electricity in households’ economy. The missing link is then the distribu-tion of electricity consumpdistribu-tion over income groups. This has not been readily available for the other Nordic countries, and we assume that the situation in the other countries is similar to that in Norway. Given the relative importance of electricity in Norway, this is probably a somewhat pes-simistic assumption.
Another issue is the share of consumers with fixed contracts. In the short run, consumers with fixed contracts are less exposed to high spot market prices, and there was clearly a great differ-ence between such exposure in Norway and the other countries in 2002/03. While consumers in the other countries mostly faced fixed price contracts, the major share of Norwegian consumers had “variable price” contracts, implying that retailers can adjust prices on a regular basis. How-ever, the analysis in [12] shows that prices of fixed contracts increased significantly towards the end of 2002. As a result, the increase in electricity expenditure between 2002 and 2003 was probably of a similar magnitude for Swedish as for Norwegian consumers. This supports the as-sumption that in the long run, spot prices have a similar impact on consumer expenses, regardless of contract form5. The only difference is that with e.g. one-year contracts, the effect is spread over a longer period, and therefore less obvious. The downside of annual contracts is reduced demand elasticity on a seasonal basis. As a result, contract forms do not have an impact on the effect of high energy prices on the economy as a whole, but they may dampen the political turmoil because the effect on households is less pronounced in the short run.
Table 3-3 gives an overview over indicators of the importance of electricity in the Nordic coun-tries. We use the word “power price” for the part of the electricity price that is related to electrical energy, as opposed to grid costs and taxes. This power price share indicates how a relative
in-5 Of course, consumers who are lucky to buy a fixed price annual contract at the optimal time will be less affected by spot price increases. However, when many consumers buy such contracts, some are lucky and some are not, and the average effect of this will be a development according to expected spot prices.
12X333 TR F5962
crease in the Elspot price will affect relative household expenditure on electricity. E.g. for Swe-den, the electricity is estimated to constitute 3.2 % of household expenditure. The power price share of this is 45 %, which means that 0.45 x 3.2 % = 1.4 % is related to power generation. This means, everything else equal, that a doubling of the spot price in a whole year would give the av-erage Swedish consumer an economic loss equal to 1.4 % of his or her expenditure. The numbers are based on 1999, which may cause some bias because electricity prices were low in Norway in that year.
Table 3-3: Indicators for the impact of high electricity prices
Sweden Norway Finland Denmark
1 Energy consumption (TWh) TWh 411.9 236.4 293.3 181.9
2 Net electricity consumption TWh 135.6 105.7 80.8 32.8
3 Net electricity consumption Share of 1) 0.33 0.45 0.28 0.18
4 Household energy consumption TWh 92.9 44.2 60.4 50.7
5 Household electricity consumption TWh 41.9 37.0 20.6 9.6 6 Household electricity consumption Share of 2) 0.31 0.35 0.25 0.29 7 Household electricity consumption Share of 4) 0.45 0.84 0.34 0.19 8 Household electricity expenditure share
of total household expenditure
% 3.2 3.0 2.3 2.4
9 Power price share of electricity price1 % 45.0 40.0 45.0 25.0 10 Household power expenditure share
of total household expenditure
% 1.4 1.2 1.0 0.6
1Approximate share, including VAT on power price
The importance of electricity in the energy mix is clearly highest in Norway, somewhat lower in Sweden and Finland and considerably lower in Denmark. Households’ share of electricity con-sumption (6) is rather similar for all countries, but households’ electricity concon-sumption as share of their energy consumption (7) is widely different. However, the estimated expenditure to the power share of the electricity bill as a share of households’ total expenditure (8) is surprisingly equal.
As argued, high prices do not present a problem to the average consumer, but to groups of low-income households with high electricity consumption. In spite of the numbers in the last line of the table above, we believe that this problem is worst in Norway, given the high share of electrical heating. On the other hand, the numbers may indicate that high prices are a greater problem than earlier assumed for households also in the other countries.
3.4.1.3 Classification of high-price situations
We now make a link between the observations of the impact of high prices and vulnerability. We have argued that the Norwegian economy absorbed the high prices in 2002/03 surprisingly well.
However, experience from Norway shows that the pressure of public opinion on the political
12X333 TR F5962
thorities can become high. This may ultimately result in measures that reduce the efficiency of the electricity market. We choose to regard the combination of the effect on low-income groups and the possibility of ill-advised intervention as an aspect of vulnerability, and measure it by the direct economic loss of Nordic consumers, defined as a price increase compared with normal prices multiplied with total consumption6.
Average consumer prices in 2003 in Norway were estimated to be 205 NOK/MWh (12.2 €/MWh) (including VAT) over a normal level, or 25 €/MWh, using the average exchange rate for March 2003-February 2004. Elspot prices did not vary much between countries in 2003. The direct eco-nomic loss to Nordic household consumers can therefore be estimated to 2.7 billion Euros includ-ing VAT, or roughly 2.2 billion Euros excludinclud-ing VAT.
The average spot price in 2003 was about 300 NOK/MWh (37 €/MWh) or roughly 100
NOK/MWh (12.2 €/MWh) higher than could be expected in a normal hydrological year. The in-crease of 205 NOK/MWh (2.5 €/MWh) to consumers in Norway is partly due to VAT, partly to retailers’ hedging costs and partly to increased profits. We assume that the damage to society is proportional to the difference between the spot price and a “normal” spot price multiplied with total consumption, which for 2003 would amount to 12.2 €/MWh multiplied with 397 TWh (gross consumption in 2002) or 4.8 billion Euros.
Based on the discussion above, the consequences of the situation in 2002/03 are judged as “Mod-erate”. We now make the following supposition: if the Elspot price would become 1 NOK/kWh in two months, it would become politically unavoidable in Norway to intervene in the market. This would have to include some form of physical curtailment7. It is arguable how such a situation should be characterized, but with increasing curtailment, there would be a critical situation. A price of 1 NOK/kWh would not occur suddenly. An illustration of an assumed critical scenario is given in Figure 3-8. In the figure the prices are given in NOK/MWh and €/MWh, a normal price of 200 NOK/MWh (24.4 €/MWh) is assumed, and the highest prices are assumed to occur in March-April. This would give an average spot price of 233 NOK/MWh ((40 €/MWh) over normal for the year as a whole, and an estimated economic loss to Nordic household consumers of 4.0 billion Euros and of 8.8 billions Euros to all Nordic consumers.
6 This is only a rough estimate. I [Halvorsen og Nesbakken] another measure, Compensating Variation, is used, which is theoretically more correct. However, the difference is not very large, and within the context of this project our es-timate is acceptable.
7 If one of the countries in the integrated Nordic market should decide on curtailment, this would be a major disrup-tion of the whole market.
0
jan feb mar apr mai jun jul aug sep okt nov des
price (NOK/MWh)
Figure 3-8: Illustration of critical price development that may lead to regulatory action and physical, non price-based curtailment
Figure 3-8: Illustration of critical price development that may lead to regulatory action and physical, non price-based curtailment
On this basis we end up with the following classification:
On this basis we end up with the following classification:
Table 3-4: Classification of High-price situations (excl VAT) Table 3-4: Classification of High-price situations (excl VAT) Direct economic loss to
< 1.0 billion Euros < 2.2 billion Euros None
1.0 – 2.5 billion Euros 2.2 – 5.5 billion Euros 25 €/MWh in one year Moderate 2.5 – 4.0 billion Euros 5.5 – 8.8 billion Euros 36 €/MWh in one year Major
> 4.0 billion Euros > 8.8 billion Euros > 36 €/MWh in one year, curtailment
Critical
Thus, in the context of energy shortage, minor events are not defined. Although it naturally would be possible to subdivide the outcomes below 2.2 billion Euros to all consumers in “Minor” and
“No consequence”, we do not think this is fruitful in the context of energy shortage and vulner-ability. Given the intention to let prices balance demand and supply, some price variation must be reckoned with, and within certain limits, this cannot be seen as an “unwanted event”. Due to the difficulties in assessing the necessary amount of curtailment and demand elasticity in prolonged periods of very high prices, we have not attempted to find a limit where shortages could be classi-fied as “Catastrophic”.
12X333 TR F5962
12X333 TR F5962